Polygraph tests, often regarded as the ultimate tool for assessing truthfulness, raise questions about their diagnostic value and reliability. Clients—whether businesses, families, or even legal institutions—frequently inquire about the accuracy of these tests. Terms like “95% accuracy” are commonly cited, but understanding the nuances of diagnostic value versus error probability is essential.
Understanding Diagnostic Value in Polygraph Tests
The diagnostic value of a polygraph test refers to its accuracy and reliability. This depends on several factors:
- Test Type: Single-issue tests (e.g., You-Phase or Utah ZCT) versus multi-issue tests (e.g., Utah MGQT or USAF MGQT v.2).
- Number of Relevant Questions: More questions can affect diagnostic precision.
- Scoring Methodology: Systems like ESS (Empirical Scoring System) or Utah scoring impact the results.
Single-issue tests generally have higher diagnostic value. For instance:
- You-Phase (Bi-Zone): 90.4% diagnostic value (ESS).
- Utah ZCT: 92.1% (ESS) or 93% (Utah scoring).
By comparison, multi-issue tests like Utah MGQT, despite allowing up to four critical questions, achieve a diagnostic value of 87.5%. This makes them unsuitable for evidentiary purposes but popular for private cases where multiple questions are prioritized.
What Is Error Probability?
Error probability, distinct from diagnostic value, refers to the likelihood of an incorrect result. It is calculated by analyzing physiological responses across multiple test series (3–5 repetitions). Each question (spot) is evaluated individually through channels such as:
- Pneumo: Breathing patterns.
- Cardio: Heart rate.
- EDA (Electrodermal Activity): Skin conductance changes.
- Plethysmograph (optional): Blood volume changes.
In single-issue tests, scores for all critical questions (e.g., R1, R2, R3 in Utah ZCT) are summed across series. A conclusive score indicates whether the subject’s responses align with truthful or deceptive patterns. Occasionally, responses are inconclusive, though this is rare.
In multi-issue tests, the lowest score among all questions determines the final evaluation. For example:
- R4: +2
- R5: +8
- R7: +1
- R8: +4
Here, the evaluation relies on R7’s score (+1), which has the highest error probability.
If the calculated error probability is 2.4%, the test result is 97.6% reliable. This is often misinterpreted as the test’s diagnostic value, underscoring the need to distinguish between the two concepts.
Real-World Applications
For evidentiary purposes, organizations like the American Polygraph Association (APA) and British Polygraph Society recommend tests with diagnostic values of 90% or higher. For private or business cases, multi-issue tests are more flexible, balancing reliability with practicality.
The Role of Neutral Environments
Test environments also influence results. For example, a recent private examination conducted in Katowice emphasized neutrality to ensure psychological comfort for both the subject and the polygrapher. While unconventional, such accommodations can enhance the reliability of results.
Conclusion
Understanding the diagnostic value and error probability of polygraph tests is key to interpreting their results. While single-issue tests provide higher accuracy, multi-issue tests offer flexibility for private cases. Clients must consider these factors to make informed decisions about the use of polygraph examinations.